Title:

Application of stochastic models to radiation chemistry

This thesis addresses one area of major interest in reaction kinetics, the theoretical description of recombination in nonhomogeneous systems. The reaction of the highly reactive particles formed by the passage of ionising radiation through a medium is an important example of this type of system. Stochastic effects are apparent in every stage of the development of a radiolysis system: in the interaction between the radiation and the medium and in the diffusion and reaction processes that follow. Conventional models for nonhomogeneous kinetics in radiation chemistry are based upon a deterministic analysis. These models are appraised together with an alternative stochastic approach. Three stochastic methods are discussed: a full Monte Carlo simulation of the diffusion and reaction and two approximate models based upon an independent pairs approximation. It is important that any kinetic model accurately describes the system it purports to represent and this can only be assured by extensive validation. The stochastic models are developed to encompass the diffusioncontrolled reactions of ions and radicals and to include the effects of a bulk electrolyte upon the reactions between ions. To model a realistic radiation chemistry reaction scheme, it is necessary to consider reactions that are not fully diffusioncontrolled. The radiation boundary condition is introduced and used to extend stochastic modelling to partially diffusioncontrolled reactions. Recombination in an anisotropic environment is also considered. Although a complete analysis of the chemistry of a radiolysis system requires a complex reaction scheme, the scheme can be simplified, in acid and in alkali, by the use of an instantaneous scavenging approximation. In acid, this approximation produces a simple three reaction mechanism based upon five species: H, OH, H_{2} , H_{2}0 and H_{2}0_{2} . The acid system is used to demonstrate the stochastic treatment of realistic kinetics. The instantaneous scavenging approximation is examined in detail and techniques are developed for the explicit modelling of reactions with a homogeneously distributed scavenger. A stochastic treatment of nonhomogeneous reaction kinetics requires a description of the initial spatial distribution of the reacting particles. A rudimentary Monte Carlo simulation is used to determine a simple distribution of clusters of reactive particles similar to that found along the path of a high energy electron in water. This distribution provides a suitable basis for kinetic simulation. The kinetics of a more detailed idealised track structure are also considered and the stochastic and deterministic kinetics of extended particle distributions are discussed.
